Raimond L. Winslow, Ph.D. - US grants

The objective of this research is to undertake a comprehensive study of the ways in which properties of single sinus and atrial cardiac cells such as voltage-dependent membrane currents, ion pumping mechanisms, calcium buffering and release systems, and inter-cellular gap junction conductance, influence the generation and propagation of electrical activity in models of the sinus node, atrium, and combined sinus node - atrial networks. Cell models will be based on whole-cell voltage-clamp data obtained from sinus and atrial cells isolated from rabbit heart. Networks will be modeled as two-dimensional lattices, with neighboring cells interconnected by resistors representing gap junctions. Network stat equations will be integrated on a massively parallel supercomputer, the Connection Machine CM-2, in order to facilitate simulation of very large systems of cells. The models will be used to study the mechanisms which influence the generation and propagation of pacemaker activity in normal cardiac tissue. Mechanisms giving rise to arrhythmias in localized groups of cells, and factors affecting the propagation of arrhythmic activity through model networks will also be investigated. The proposed research will enhance our basic understanding of both normal and abnormal cardiac dynamics, and is unique in that it represents an attempt to understand the "macroscopic" properties of large networks of cardiac cells give detailed descriptions of the "microscopic" behavior of individual cells.

Winslow 9404844 The objective of this research is to utilize modern principles of nonlinear dynamical system theory in conjunction with the power of massively parallel computing to study, in great biophysical detail, the effects of ischemia on the dynamics of single cardiac cells and networks. It will use a physiologically detailed model of single cardiac cells which consists of 16 coupled ordinary differential equations. Tools for chaos theory will then be used to analyze the model.

The goal of this project is to use large-scale, biophysically detailed models of the mammalian sinus node and atrium to understand dynamics of interaction between these two systems under a variety of conditions. Among the specific aims are: 1) development of an SA node - atrial network model incorporating recently discovered date on the interdigitation of these two tissue regions; 2) development of a network model of vagal control of heart rate. Model studies will involve a combination of single cell and network analyses. Dynamics of single cells will be studied using techniques of nonlinear dynamical systems theory. Network simulations will be implemented on both the parallel Connection Machine CM-5 and the symmetric multi-processing Silicon Graphics Power Challenge XL supercomputers.

DESCRIPTION (Adapted from investigator's abstract): Congestive heart failure is a primary cardiac disease that affects roughly 1% of the US population. Mortality in the first five years subsequent to diagnosis ranges from 35-60%. The primary cause of mortality during this period is a severe electrical arrhythmia known as Sudden Cardiac death (SCD). the cuases of SCD are unknown. The objective of this resarch is to undertake a joint modeling and experimental study directed at developing quantitative computer models of electrical excitation, propagation, and repolarization in the failing heart in order to better understand the origins and prevention of SCD. The canine tachycardia pacing-induced animal model of heart failure is used, as this model yields hemodynamic and electrical changes in the canine heart which are strikingly similar to those seen in human patients. Aim 1 will formulate a computer model of the normal canine isolated ventricular cell action potential, and use this model in conjunction with experiments to examine dependence of action potential shape and duration on intracellular calcium handling processes, and repolarizing membrane currents. Aim 2 will develop a computer model of the failing canine ventricular myocyte and will use the model to investigate whether or not the altered expression of repolarizing membrane currents and proteins involved in intracellular calcium handling known to occur during heart failure can account for action potentials and intracellular calcium transients measured in failing cells. Aim 3 will undertake a combined modeling and experimental study investigating three possible sources of arrhythmia in failing cells: a) early afterdepolarizations; b) oscillatory pre-potentials; and c) altered expression of the If pacemaker current. Aim 4 will use diffusion tensor magnetic resonance imaging to measure changes in anatomical structure of normal versus failing canine hearts. These structural data will be used in conjunction with the models developed in Aims 1-3 to investigate electrical excitation, propagation, and repolarization in three-dimensional models of the failing canine ventricles. These computer models will also be used to investigate the ways in which ventricular dilatation, wall thinning, and possible alteration of fiber orientation and/or fiber rotation gradient alter electrical conduction in the failing heart. The arrhythmogenic potential of the cellular mechanisms investigated in Aim 3 will be tested using the three-dimensional ventricular models.

DESCRIPTION (Adapted from investigator's abstract): Congestive heart failure is a primary cardiac disease that affects roughly 1% of the US population. Mortality in the first five years subsequent to diagnosis ranges from 35-60%. The primary cause of mortality during this period is a severe electrical arrhythmia known as Sudden Cardiac death (SCD). the cuases of SCD are unknown. The objective of this resarch is to undertake a joint modeling and experimental study directed at developing quantitative computer models of electrical excitation, propagation, and repolarization in the failing heart in order to better understand the origins and prevention of SCD. The canine tachycardia pacing-induced animal model of heart failure is used, as this model yields hemodynamic and electrical changes in the canine heart which are strikingly similar to those seen in human patients. Aim 1 will formulate a computer model of the normal canine isolated ventricular cell action potential, and use this model in conjunction with experiments to examine dependence of action potential shape and duration on intracellular calcium handling processes, and repolarizing membrane currents. Aim 2 will develop a computer model of the failing canine ventricular myocyte and will use the model to investigate whether or not the altered expression of repolarizing membrane currents and proteins involved in intracellular calcium handling known to occur during heart failure can account for action potentials and intracellular calcium transients measured in failing cells. Aim 3 will undertake a combined modeling and experimental study investigating three possible sources of arrhythmia in failing cells: a) early afterdepolarizations; b) oscillatory pre-potentials; and c) altered expression of the If pacemaker current. Aim 4 will use diffusion tensor magnetic resonance imaging to measure changes in anatomical structure of normal versus failing canine hearts. These structural data will be used in conjunction with the models developed in Aims 1-3 to investigate electrical excitation, propagation, and repolarization in three-dimensional models of the failing canine ventricles. These computer models will also be used to investigate the ways in which ventricular dilatation, wall thinning, and possible alteration of fiber orientation and/or fiber rotation gradient alter electrical conduction in the failing heart. The arrhythmogenic potential of the cellular mechanisms investigated in Aim 3 will be tested using the three-dimensional ventricular models.

Purpose: Core E (the Modeling Core), functions as a resource for the individual projects by providing expertise in: a) the design of mathematical models of cardiac ion channels, membrane currents, intracellular Ca handling processes, cells, and tissue; b) efficient software implementation of these models on scalar and parallel computers; and c) model-based analyses of experimental data and experimental design.

We are requesting support to purchase an IBM SP-3 Night Hawk 1 parallel computer. This computer will be used by a core group of 5 major users at the Johns Hopkins University School of Medicine and Whiting School of Engineering. A total of 10 projects are proposed. These projects are supported by a total of ten RO1, one R37, and one PO1 NIH awards. Three of the major users are principal investigators on four of these awards. One major user is a Co-Principal investigator on four of these awards. The scope of the projects are broad, and include: a) an experimental and modeling study of the basis of arrhythmia in heart failure; b) computational modeling of cardiac sodium channel structure and function; c) computational modeling of metabolic oscillations in cardiac ventricular cells; d) computational modeling of blood flow and molecular transport in the microcirculation; e) computational modeling of mechanics and electromotility of the cochlear outer hair cell; f) computational mapping of brain structure and function. These projects have several features in common. First, each modeling study is linked closely to the underlying biology of the system being studied. Second; because the models being developed incorporate a high degree of biophysical and structural detail, they are computationally demanding. Consequently, each of the researchers in this proposal have previously used parallel computing to solve their respective problems. However, problem size has grown sufficiently that existing resources can no longer meet the needs of this group. Award of this instrument will have an immediate and direct stimulating effect on each of these projects. It will also enhance the ability of the Whitaker Biomedical Engineering Institute at Johns Hopkins University and the Department of Biomedical Engineering to play a leadership role in the field of computational modeling of complex biological systems at the local, national, and international levels.

DESCRIPTION (taken from the application): The amount and rate of accumulation of biological information is increasing rapidly, as is demonstrated by the fact that there are now over 300 biological databases accessible from the World Wide Web. There is growing recognition, however, that the emergent, integrative behaviors of biological systems are a result of complex dynamic interactions between all the components from which these systems are composed, and that knowledge of each system component, however detailed, is not sufficient by itself to understand these integrative behaviors. It is our thesis that a quantitative understanding of biological function will only be achieved through development of structurally, biochemically and biophysically detailed computational models that are based directly on experimental data. Once developed, these models can be simulated, analyzed and understood through application of modem engineering and computational approaches. We therefore propose to create a Consortium for Integrative Modeling of Biological Systems. The goal of this Consortium will be to develop and apply computational models to better understand functional properties of biological systems at multiple structural and functional levels ranging from that of regulation of gene expression to that of the cell, and in some instances the organ level. This Consortium will develop new approaches for the generation, storage and dissemination of computational models, and web-based course materials in support of education and research training in computational modeling at undergraduate, graduate, medical, postdoctoral and general scientist levels.

Dilated cardiomyopathy (DCM) is the most common form of primary cardiac muscle disease, with prevalence estimated at 36.5 cases per 100,000. DCM is characterized by ventricular dilation, decreased myocardial contractility and cardiac output, and increased risk of sudden cardiac death. Ventricular myocytes isolated from failing hearts exhibit changes in expression levels of proteins involved in repolarization of the action potential (AP) and intracellular calcium (Ca2+) cycling. These changes are accompanied by reduction of junctional sarcoplasmic reticulum (JSR) Ca2+ concentration, peak intracellular Ca2+ transient amplitude, slowed diastolic Ca2+ extrusion and prolongation of AP duration. We have previously formulated a "minimal" computational model of the failing canine ventricular myocyte that incorporates experimental data on down-regulation of potassium (K+) currents and the SR Ca2+-ATPase, and up-regulation of the Na+-Ca2+ exchanger. This model is able to qualitatively reconstruct changes in AP and Ca2+ transient morphology observed in failing myocytes. Model simulations predict that down- regulation of the SR Ca2+-ATPase by itself produces significant prolongation of AP duration by reducing JSR Ca2+ level, JSR Ca2+ release and the magnitude of Ca2+-dependent inactivation of L-type Ca2+ current (ICa,L). This decreased Ca2+-dependent inactivation increases ICa,L during the plateau phase, thereby increasing AP duration. These model predictions are supported by results of preliminary experiments. This has led us to hypothesize that JSR Ca2+ level through effects on JSR Ca2+ release and Ca2+-dependent inactivation of ICa.L, modulates AP duration, and that this modulation is important under a range of conditions producing changes in JSR Ca2+ level, including heart failure. The general goal of the proposed research is to test this hypothesis by means of experiments coupled with computational modeling.

[unreadable] DESCRIPTION (provided by applicant): The fiber structure of the heart plays a critical role in shaping electrical propagation. Conduction is influenced by tissue geometric factors such as expansion and contraction, and is anisotropic, with current spread being most rapid in the direction of the fiber long axis. Spatial rate of change of fiber orientation also influences conduction properties. Remodeling of ventricular geometry and fiber organization, including development of interstitial fibrosis, is a prominent feature of several cardiac pathologies, and these alterations may figure importantly in arrhythmogenesis. A detailed knowledge of ventricular fiber structure, how it may be remodeled in cardiac pathology, and the effects of this remodeling on ventricular conduction is therefore of fundamental importance to the understanding of cardiac electro-mechanics in health and disease. We will investigate how anatomical remodeling of ventricular fiber structure influences ventricular conduction, using the canine tachycardia pacing-induced heart failure preparation as a model system. Several aims must be accomplished to do this. First, we will develop MR imaging methods for the rapid reconstruction of ventricular fiber structure. Second, we will use these methods to measure fiber structure in populations of normal and failing hearts. Third, we will develop mathematical methods for identifying statistically significant changes in fiber structure between normal and failing hearts. Fourth, we will measure electrical activation patterns in each heart that is anatomically reconstructed using MR imaging methods. Fifth, we will relate measured changes in fiber structure to measured changes of electrical propagation in each heart using both experimental approaches as well as computational models.

This is designed to investigate how mitochondrial energetics and integrated cardiac function are remodeled by ischemia and reperfusion in a comprehensive manner, spanning from the molecular level, to the proteome, to global effects on excitation-contraction (EC) coupling and oxygen consumption in intact muscle. However, as a result of complex control interactions within the machinery of oxidative phosphorylation and between mitochondria, the sarcolemma and the cytoplasm, it is difficult to form a complete mental picture of how a particular experimental finding impacts myocyte function as a whole. To help address this challenge, the COMPUTATIONAL/BIOINFORMATICS will develop integrative computational models of the cardiac ventricular myocyte incorporating biophysically detailed descriptions of mitochondrial energetics, which will be used to interpret results from the range of experiments being pursued in this project. Our team has pioneered the development of detailed models of the electrophysiology and Ca2+ handling properties, ion transport processes and mitochondrial energetics, which we are now integrating into comprehensive virtual cardiac ventricular myocyte models. The emphasis has been on using these models as tools for the interpretation of experimental results and for suggesting new experiments that can reveal the fundamental nature of the control of myocyte function under normal or pathophysiological conditions. We will focus the existing collaboration of the leaders on the problem of mitochondrial dysfunction in the post-ischemic heart using a common experimental animal model and a common quantitative mathematical/computational model. It will also foster new collaborative interactions between experts in diverse disciplines. In addition to developing and exploiting the computational model, we will provide expertise and support in the quantitative methodology of metabolic control analysis, for identifying the key controlling factors in oxidative phosphorylation in the post-ischemic heart.

DESCRIPTION (provided by applicant): We are proposing to establish the Cardiovascular Research Grid (CVRG). The CVRG will provide the national cardiovascular research community a collaborative environment for discovering, representing, federating, sharing and analyzing multi-scale cardiovascular data, thus enabling interdisciplinary research directed at identifying features in these data that are predictive of disease risk, treatment and outcome. In this proposal, we present a plan for development of the CVRG. Goals are: To develop the Cardiovascular Data Repository (CDR). The CDR will be a software package that can be downloaded and installed locally. It will provide the grid-enabled software components needed to manage transcriptional, proteomic, imaging and electrophysiological (referred to as "multi-scale") cardiovascular data. It will include the software components needed for linking CDR nodes together to extend the CVRG To make available, through community access to and use of the CVRG, anonymized cardiovascular data sets supporting collaborative cardiovascular research on a national and international scale To develop Application Programming Interfaces (APIs) by which new grid-enabled software components, such as data analysis tools and databases, may be deployed on the CVRG To: a) develop novel algorithms for parametric characterization of differences in ventricular shape and motion in health versus disease using MR and CT imaging data;b) develop robust, readily interpretable statistical learning methods for discovering features in multi-scale cardiovascular data that are predictive of disease risk, treatment and outcome;and c) deploy these algorithms on the CVRG via researcher-friendly web-portals for use by the cardiovascular research community To set in place effective Resource administrative policies for managing project development, for assuring broad dissemination and support of all Resource software and to establish CVRG Working Groups as a means for interacting with and responding to the data management and analysis needs of the cardiovascular research community and for growing the set of research organizations managing nodes of the CVRG. (End of Abstract).

DESCRIPTION (provided by applicant): Sudden Cardiac Death (SCD) remains a leading cause of death in the western world. Estimates suggest that roughly 10-20% of all annual mortality in the U.S. results from SCD and that approximately 5% of the middle-aged U.S. population has a significant predisposition to SCD. The major causes of SCD in adults age 35 and older are coronary artery disease (CAD;~ 80%) and dilated cardiomyopathy (~10-15%), with risk increasing dramatically with age. While implantable cardioverter defibrillators (ICDs) are proving to be effective in reducing the occurrence of SCD, wholesale deployment of ICDs in large patient populations is impractical economically and ignores the facts that the majority of patients with ICDs are likely never to require them and that there are as yet no effective means for identifying patients at highest risk for SCD. Working both from the top-down (whole heart optical mapping, anatomical reconstruction and simulation) and from the bottom up (mitochondrial and cellular imaging and modeling), our aim is to achieve an unprecedented level of integration of structure and function in order to understand and model the ways in which coupling between metabolic and electrophysiological processes in the myocyte contribute to risk of cardiac arrhythmias under conditions of metabolic stress. We refer to this as the metabolic sink hypothesis. Cluster Project 1 will test the hypothesis that metabolic sinks may be formed by producing local regions of IKATP activation in the intact-perfused guinea pig (GP) heart and will assess their impact on ventricular conduction and arrhythmia generation. Cluster Project 2 will test the hypothesis that metabolically stressed myocardium is particularly susceptible to formation of metabolic sinks leading to arrhythmia in the setting of heart failure. Cluster Project 3 will develop novel biophysically, metabolically and anatomically detailed computational models of electrical conduction and, in conjunction with Cluster Projects 1 &2, test hypotheses regarding the ways in which the interplay between metabolic and electrophysiological function contributes to generation of arrhythmias under conditions of metabolic stress. The metabolic sink hypothesis has never been tested directly by producing metabolic uncoupling of mitochondria in local regions of myocardium and measuring effects on electrical conduction and generation of arrhythmias. Whether or not failing myocytes are susceptible to metabolic oscillations, whether or not failing tissue is more or less susceptible to formation of metabolic sinks than is normal tissue, and whether or not metabolic sinks form a substrate for reentry in failing myocardium is unknown. This project will test these hypotheses and the results will have major importance for our understanding of the mechanisms and treatment of arrhythmias.

[unreadable] DESCRIPTION (provided by applicant): [unreadable] [unreadable] The short course "Integrative Models of the Cardiac Myocyte" will train biomedical researchers and clinician-scientists in how to develop and apply biophysically- and biochemically-detailed, experimentally-based computational models of the cardiac myocyte. Emphasis will be on modular development and testing of models using experimental data, numerical methods for solving model equations, mathematical techniques for developing and solving multi-scale models of the cell and methods for analyzing parameter sensitivity of models. The approach to this course is that developed by Dr. Harry Goldberg and which is being incorporated extensively into the JHU School of Medicine curriculum. In this approach, cd-rom and/or web-based learning modules are developed and provided to students prior to lecture. These learning modules present the core material for each lecture. Lectures are then used as an opportunity for direct interaction with students, to address questions on the core material and to apply or delve more deeply into the core material. These learning modules will be developed using Macromedia Breeze and will be available to everyone via the web. Instructional materials on how to create new learning modules will be created, thus enabling the cardiovascular research community to download existing learning modules, and generate new learning modules of their own design which may be added to their instantiation of the web course in order to meet local educational needs. [unreadable] [unreadable] [unreadable]

[unreadable] DESCRIPTION (provided by applicant): This shared equipment grant requests funds to acquire a 256 dual-quad node, large memory cluster computer. This computer will become the major shared computing resource used by faculty, trainees, and staff of the Institute for Computational Medicine (ICM) at Johns Hopkins University. This computer will support disease-related research in three major areas: a) modeling of biological systems; b) computational anatomy; and c) mathematical bioinformatics. Research in biological systems is directed at understanding the mechanisms of Sudden Cardiac Death (SCD; the major cause of death in the western world) using combined experimental and modeling approaches. This work is providing novel insights into the molecular and cellular mechanisms of SCD, and is enabling the "in-silico" design of optimal approaches for terminating life-threatening arrhythmias using shocks applied to the heart by implantable cardioverter defibrillators (ICDs). ICM research in computational anatomy (CA) is directed at developing algorithms for discovering changes in the anatomy and function of the brain, as well as changes in the structure/function and motion of the heart, that predictive developing brain or heart disease. Structure, function, and/or motion is measured in both normal and disease populations by analyzing three-dimensional image volumes acquired [unreadable] from patients using magnetic resonance imaging or computed tomography. CA algorithms are then used to discover changes in anatomical structure, function, and/or motion that distinguish normal and diseased populations with high accuracy. These methods are used for the early diagnosis of developing brain or heart disease so that therapeutic interventions can be made. Major application areas are the discovery of features that signal early onset of diseases such as schizophrenia, dementia of the Alzheimer's type, ischemic versus idiopathic dilated cardiomyopathy, and risk of SCD. Research in mathematical bioinformatics is directed at developing novel computational algorithms that operate on patient-specific multi-scale data (e.g., data on single nucleotide polymorphisms, transcript levels, protein expression levels, imaging data, and clinical data) to discover biomarkers for accurate, sensitive, and specific prediction of disease onset, stage, risk and therapeutic approach. Applications of these computational methods are broad, including discovery of cancer biomarkers and early prediction of risk for SCD so that ICD placement therapy may be performed. Achieving each of these goals requires the development and application of computer models and algorithms that are very [unreadable] computationally intensive. The present computing resources of the ICM are no longer state of the art, and research progress is being slowed. PUBLIC HEALTH REVELANCE: Award of the requested instrument will dramatically accelerate the development and application of computational models and algorithms for the early diagnosis and treatment of brain disease, heart disease, and cancer. [unreadable] [unreadable] [unreadable]

1,4,5-IP3; 1,4,5-InsP3; ATP phosphohydrolase (Na+ K+ transporting); Anatomic structures; Blood Coagulation Factor IV; CRISP; Ca Release Channel-Ryanodine Receptor; Ca++ element; Calcium; Calcium-Ryanodine Receptor Complex; Cardiac; Cardiac Myocytes; Cardiocyte; Cell Communication and Signaling; Cell Signaling; Coagulation Factor IV; Computer Retrieval of Information on Scientific Projects Database; D-myo-Inositol, 1,4,5-tris(dihydrogen phosphate); Data; Electrons; Experimental Models; Experimental Models, Other; Factor IV; Funding; Grant; Heart myocyte; Image; Inositol 1,4,5-Triphosphate; Inositol 1,4,5-Trisphosphate; Institution; Intracellular Communication and Signaling; Investigators; Ion Channel; Ionic Channels; Location; Maps; Membrane Channels; Methods and Techniques; Methods, Other; Microscopic; Mitochondria; Modeling; Models, Experimental; Muscle Cells, Cardiac; Muscle Cells, Heart; Myo-Inositol 1,4,5-Trisphosphate; Myocytes, Cardiac; Myoinositol 1,4,5-Triphosphate; NIH; Na(+)-K(+)-Exchanging ATPase; Na(+)-K(+)-Transporting ATPase; Na+ K+ ATPase; National Institutes of Health; National Institutes of Health (U.S.); Negative Beta Particle; Negatrons; Organelles; Placement; Potassium Pump; Property; Property, LOINC Axis 2; Receptor Protein; Research; Research Personnel; Research Resources; Researchers; Resources; Role; Ryanodine Receptor; Ryanodine Receptor Calcium Release Channel; Sarcoplasmic Reticulum; Signal Transduction; Signal Transduction Systems; Signaling; Sodium Pump; Sodium, Potassium ATPase; Sodium, Potassium Adenosine Triphosphatase; Sodium, Potassium Adenosinetriphosphatase; Sodium-Potassium Pump; Source; Structure; Study models; Techniques; United States National Institutes of Health; base; biological signal transduction; cardiomyocyte; experiment; experimental research; experimental study; imaging; mitochondrial; nano meter; nanometer; receptor; research study; social role; sodium potassium exchanging ATPase; tool

DESCRIPTION (provided by applicant): The structure of tissues (such as heart or skeletal muscle) undergoes substantial remodeling in disease, including macroscopic changes, e.g.: overall geometry;vascular network reorganization;fiber and sheet orientation;and interstitial fibrosis;and microscopic changes, e.g.: redistribution of molecular expression;and the energetic state of mitochondria. There is an urgent need to understand the nature of this remodeling, because all of these processes impact on the function of these tissues in health and disease, but existing methods for tissue reconstruction are low-resolution, while current high-resolution sub- cellular imaging cannot be expanded to tissues and whole organs. Here we describe an unprecedented approach by which two-photon microscopy will be used to perform whole-tissue 3D reconstructions at micron- level spatial resolution, bridging the gap from micro to macro scale. We request a state-of-the-art Zeiss 7MP dedicated multiphoton microscope system for the multiphoton- excited fluorescence imaging and automated 3D reconstruction of tissues, to be used by multiple NIH-funded investigators to accomplish research not currently possible with existing equipment. Integration of a motorized stage and automated microtome allow this instrument to produce images at a higher resolution, for a larger volume of tissue, than has been yet achieved, making three-dimensional reconstructions of large-tissue- volume images possible. The system will initially have 7 major users and will be fully utilized for long imaging sessions of large tissues;as research evolves, minor users may be added. This will be the only instrument of its kind at the Johns Hopkins University. All other multiphoton microscopes at the university are used for different applications and cannot be used or modified for our purposes;this is the only one at JHU to be used for these time-consuming tissue reconstructions. The University has identified this need and backs our proposal with financial support for infrastructure to accommodate this imaging system. Relevance: The proposed system removes roadblocks to a wide range of currently NIH-funded research, and opens the door to a breadth of future work of significant impact on human health. Applications include: atlases of healthy and diseased hearts in multiple species;and computational analysis of both the population-level and individual-level anatomical changes that occur. Many of the downstream applications involve using the high- resolution datasets generated to create complex anatomically-detailed and individualized computational models of cardiac electrophysiology, electromechanics, blood flow and drug delivery, to simulate Sudden Cardiac Death, Cardiac Defibrillation, Peripheral Arterial Disease and more. In fact, we envision most data acquired being used at least three times: as an image;as a component of an atlas;and as the basis of at least one computational model.

DESCRIPTION (provided by applicant): Heart failure is a disease that is continually increasing in prevalence worldwide. In the United States, nearly 6 million people suffer from heart failure and it is the most common inpatient diagnosis in the elderly. The economic impact for 2009 has been estimated at $37.2 billion. Treatment of this disease with 2-blockers and/or inhibitors of renin-angiotensin signaling has decreased mortality and morbidity over the years, but mortality still approaches 60% within 5 years of diagnosis. Fatal arrhythmias, known as Sudden Cardiac Death (SCD), account for about half of the early deaths in HF, with progressive cardiac decompensation accounting for the remainder. Many factors contribute to the pathology of HF, including changes in the neurohumoral environment, alterations in ion channel and transporter activity, modulation of cell death pathways, and remodeling of the inherent structure of the tissue. Recent evidence indicates that alterations in the reduction- oxidation (redox) potential of the cytoplasm, sarcoplasmic reticulum, and the mitochondria of the heart may be a key factor involved in the progression of cardiac hypertrophy and failure. In heart failure (HF), there is evidence that oxidative stress may contribute to impaired function, and this may arise as a consequence of altered ion homeostasis, energetic deficiencies, and post-translational modification of protein targets. Moreover, a large number of ion channels, transporters, and signaling pathways have been shown to be modulated either directly by reactive oxygen species (ROS), or by changes in the thiol status or redox carrier concentration. Some, or many, of these targets, could contribute to an enhanced susceptibility of the failing heart to arrhythmogenesis and SCD. A comprehensive view of how shifts in metabolism and redox balance influence the electrophysiological substrate requires a systems biology approach to the problem, involving deconstruction of how individual ion channels, transporters and signaling pathways are affected by redox modulators, and how the performance of the integrated system is changed. Specifically, in this proposal, our objective is to examine how enhanced oxidative stress alters the electrophysiology, Ca2+ regulatory processes, and arrhythmia susceptibility of myocytes from failing hearts (pressure-overload model). An iterative, experimental/computational systems biology approach combining both horizontal and vertical integration will be taken. These approaches will be used to build biophysically-detailed cellular and whole-heart models of redox/antioxidant pathways and their downstream effects on ion channels and transporters, with the goal of defining how metabolic and oxidative stress leads to arrhythmias, pump failure, and SCD. An overriding goal will be to define the specific alterations that have the greatest influence on whole heart function, so as to narrow down the number of targets to pursue for therapeutic intervention.

DESCRIPTION (provided by applicant): Calcium (Ca2+) dependent arrhythmias have been identified as a significant health problem leading to ventricular tachycardia, fibrillation and death. The exact mechanism by which these arrhythmias arise remains a critically important yet vexing problem. We hypothesize that these Ca2+ dependent arrhythmias occur through a process in which a propagating wave of elevated calcium travels through heart cells and thereby activates and entrains electrical activity. The proposals PIs (Lederer, Jafri, and Winslow) will combine state-of- the art computational modeling with novel laboratory experiments in a multi-scale approach to determine how the calcium signaling defect develops and critically test the hypothesis. This systems biology investigation will examine the molecular physiology of cardiac Ca2+ signaling at high temporal and spatial resolution under normal and pathological conditions. It will utilize the unusually powerful approach of specifically examining the molecular pathophysiology of the Ca2+ dependent arrhythmia using the molecular disease, catecholaminergic polymorphic ventricular tachycardia (CPVT) caused by an extremely well-defined process - point mutations of critical Ca2+ regulatory proteins. We will examine how mutations in the calcium release channel (ryanodine receptor type 2, RyR2) and the Ca2+ binding protein, calsequestrin (CASQ2) contribute to Ca2+ dependent arrhythmogenesis. Mouse models of these two arrhythmias will be used to enable advanced cell biology investigations. The work will be made more general by also including an examination of Ca2+ overload arrhythmias in mouse and guinea pig. The planned investigation will encompass multiple scales: from the molecular defect, to cellular Ca2+ dysfunction to tissue arrhythmia using both mathematical modeling and biological experiments. The project will address the following four specific aims: 1) How do Ca2+sparks trigger and sustain calcium waves? 2) How do Ca2+waves propagate from cell to cell? How do Ca2+waves entrain electrical activity? 3) How do specific mutations in RyR2 and CASQ2 affect Ca2+sparks, Ca2+waves and the propagation of Ca2+waves from cell to cell? 4) How does the 3D organization of the heart affect Ca2+ entrained arrhythmogenesis? This work should provide fundamental new understanding of the heart and the role of Ca2+ in electrical dysfunction and arrhythmia and lay the foundation for new therapeutic approaches. PUBLIC HEALTH RELEVANCE: Calcium dependent cardiac arrhythmias have been identified as a significant health problem and a major cause of cardiac death. The proposed work will provide new understanding of the molecular and cellular causes of these heart rhythm disturbances and enable new treatments.

DESCRIPTION (provided by applicant): Electrophysiology is a branch of physiology that pertains to the electrical activity of cells and tissues, and the electrical recording techniques that enable measurement of this activity. Electrophysiological data are typically collected using single-electrodes, electrode arrays, and fluorescent imaging systems. They are the most common type of data collected in basic studies of cardiac cells and tissues, as well as many other types of excitable cells. The reason for this is that these data can be used to relate molecular function to the electrical activity of cardiac muscle cells in health and disease. Establishing this relationship is a critically important task, since sudden cardiac death (SCD) from electrical arrhythmia is now the leading cause of mortality in the Western world, exceeding cancer in deaths per year. The number of cardiac electrophysiological (CEP) studies supported by the NHLBI, and the quantity and variety of data collected in these studies of heart function, vastly exceeds that for genetic, transcriptional or proteomic studies. While there are resources (e.g., dbGaP, dbSNP, GEO, ArrayExpress, World-2DPAGE Repository, etc) for the documentation and dissemination of these latter data types, no infrastructure exists for managing and sharing celular, tissue, and whole-heart electrophysiological data. The lack of this infrastructure means that these data are being lost in the sense that they exist only in the labs of those who collect it. They are seldom, if ever, disseminated by any means other than publication in peer-reviewed journals. Published data sets are limited to a few ideal examples presented in a way (images) that does not support re-use and further quantitative analysis. To address this problem, our goal is to create software tools for annotating, storing, sharing, and querying CEP data, including both primary data from measurement instrumentation and metadata regarding experimental protocols and results of data analyses. The Cardiovascular Research Grid Project will host a national repository for these experimental data and metadata. The creation of this national repository will, for the first time, make it possible for researchers to organize, archive, and search their own data and share it with others. It will enable re-use of the data so that other groups can mine the data for new results, use the data to design new experiments, and to formulate computational models of cell, tissue and heart function in health and disease. The ability to annotate data according to species, disease, and animal models of disease (including de-identified data from human tissue), as well as other ways, will support deep analysis of the electrophysiological basis of heart disease. Each of these capabilities will help provide fundamentally important insights into the mechanisms of cardiac arrhythmias, and possible therapeutic approaches that can help reduce risk of SCD.

? DESCRIPTION (provided by applicant): The canonical chemical synapse of the nervous system is characterized by electron-dense thickening of the post-synaptic membrane in close apposition to the pre-synaptic active zone. Quite distinct and far less well- understood are the so-called c-synapses named for the sub-synaptic cistern in the post-synaptic cell that is aligned with the pre-synaptic terminal. Although common on motor neurons and cerebellar Purkinje cells among others, essentially nothing is known about c-synapse function except that they are formed by cholinergic inputs. Chief among the mysteries is the role of the subsynaptic cistern itself, although its similarity to sarcoplasmic reticulum has prompted suggestions of involvement in calcium (Ca2+) signaling. The goal of this proposal is to advance understanding of c-synapse function through a combination of computational modeling coupled with experiments on the cholinergic c-synapse. A major impediment to study of c-synapses on central neurons is the fact that there is typically no means to activate them selectively out of th multitude of inputs. We will take advantage of an exemplar c-synapse that is experimentally tractable - efferent cholinergic neurons that project from the brainstem medial olivocochlear nucleus to inhibit cochlear outer hair cells (OHCs). Because this is its sole synaptic input, the OHC will provide unique insights into how c-synapses operate. First and foremost is to determine whether the cistern regulates postsynaptic Ca2+. Remarkably, c- synapses share common design features with the fundamental structural units of excitation-contraction coupling in the heart known as dyads. In each case, a Ca2+ store (junctional sarcoplasmic reticulum JSR in heart; cistern in the OHC) is positioned close (~14 nm) to the cell membrane, creating a restricted Ca2+ nano- domain. Ca2+ sources in the cell membrane (voltage-gated Ca2+ channels in heart; nicotinic cholinergic receptors nAChRs in OHCs) direct their flux into this restricted space. In OHCs, Ca2+-activated potassium type- 2 (SK2) channels are located near the nAChRs. Even small fluxes directed into the cleft can create large Ca2+ signals that are highly localized in space and time. Ryanodine-sensitive, Ca2+-binding Ca2+-release channels (RyRs) are thought to reside in the closely apposed cistern membrane as in JSR, suggesting that the process of Ca2+-induced Ca2+-release (CICR) may be important to OHC function, as it is in the heart. We will leverage these similarities to harness extensive modeling work done in the Winslow lab on CICR in the heart, adapting these models and applying them to advance our understanding of the function of c-synapses in the nervous system. Modeling work will be informed by experiments conducted in the Fuchs lab that will probe the structure and function of the OHC c-synapses. This unique combination of an experimentally-tractable system, along with the modeling and experimental expertise of these two labs, will enable us to advance our understanding of the function of c-synapses in the nervous system.

Critical care units are the highest mortality units in any hospital. These severely ill patients undergo multiple complex interventions at the same time, and care is so complex that they are extremely vulnerable to medical errors and adverse outcomes. Critical care patients are the most heavily instrumented patients in the hospital. Physiological signals are collected using many different types of sensors. These sensor signals reflect the underlying dynamic, integrated physiological state of the patient and are thus highly complex and inter-related. The biggest challenge faced by critical care physicians is that the amount and complexity of these data push the limits of what they can cognitively assimilate and relate to the overall physiological status of their patients. They are confronted by Big Data at every moment, yet they must interpret and act on them quickly. They lack the tools to do this. In this project, a goal is to develop and apply a novel class of algorithms, known as optimal change-point detection algorithms, to the problem of detecting when a patients state changes from one clinical condition to another. The application of sophisticated algorithms to healthcare will not only transform treatment in critical care units, but will be brought to the classroom to undergraduate students minoring in Computational Medicine.

The research team plans to develop automated computational methods for processing physiological time series data from critical care patient sensors to quickly detect changes in clinical state. These methods will include; (i) improved selection of features that characterize patient state; (ii) optimal control algorithms to detect transitions of patient state based on these features. One innovation in the proposed approach is a re-formulated the state transition detection problem as an optimal change-point problem from the fields of controls-theory and Bayesian optimal sequential decision making. This in turn has enabled us to derive an optimal detector by first defining a cost function that reflects performance goals (e.g. maximize sensitivity, minimize false positives), and then developing the detection rule that minimizes this cost. The researchers have demonstrated that this approach decreases time to detection of epileptic seizure onset by 50% relative to other state of the art methods. Another innovation lies in how to compute features from physiological waveforms. The goal of proposed research is to determine if optimal-change point algorithms in conjunction with these new features applied to the analysis of multiple types of physiological time series data can support earlier detection of changes in the clinical conditions of patients in the critical care unit. A new course or course module that introduces statistical and mechanistic model estimation and simple early detection algorithms will be developed. This course will expose biomedical engineering students to experimental data, biophysical-based models, and statistical models of biological systems that have clinical relevance.

Project Summary We will create the Johns Hopkins University ?Pre-Doctoral Training Program in Computational Medicine?, in the research area of Bioinformatics and Computational Biology. The participating departments are Biomedical Engineering and Applied Mathematics & Statistics. We are requesting 6 pre-doctoral training slots per year. Our trainees will learn how to: develop models of biological systems in health and disease; constrain these models using data collected from patients; apply models to deliver improved diagnoses and therapies. They will learn this through a combination of focused course work, and dissertation research in the laboratories of program faculty. They will gain experience with oral presentation, grant writing and teaching. They will receive responsible conduct of research training that goes beyond current offerings to meet the needs of computational medicine researchers. They will be part of a community of students and faculty. We will put in place novel programs that increase diversity in training by creating an applicant pool of undergraduates who have done substantial research rotations with our program faculty. Graduates will be well prepared and sought after to fill the growing need for researchers trained in computational medicine in both industry and academia. Support of our program will contribute to achieving the goal of precision medicine.

The novel Coronavirus (COVID-19) is one of four infectious diseases caused by the SARS-CoV-2 virus. Although the clinical signs and patient symptoms of this complicated disease vary in presentation and severity, clinicians and investigators have reported constitutional symptoms (cough and fever), upper and lower respiratory tract symptoms, as well as gastrointestinal symptoms. Among the most concerning is the life threatening acute respiratory distress syndrome (ARDS) in patients. The pathophysiology of severe ARDS results from a rapid decline in pulmonary function and requires intubation of patients in critical condition for invasive mechanical ventilation to combat lung recruitability, reduced peripheral capillary oxygen saturation (SpO2) and risks of organ failure and death. Ventilator settings to increase SpO2 and oxygen delivery is achieved with positive end-expiratory pressure (PEEP). However, controlling ventilation at a high PEEP for extended periods of time significantly increases risk for ventilator-associated lung injury (VALI). This RAPID project will develop novel engineering strategies for optimal ventilator control to maximize SpO2 in minimal time, while minimizing PEEP and the duration of ventilator use are needed to minimize VALI and subsequent complications, and to improve favorable patient outcomes. In the management of patients with COVID-19, these strategies are significant to optimize oxygen delivery, minimal invasive ventilator use and mechanical lung injury. Further, the understanding of ventilator requirements and operative settings highlights the need for available ventilators. The management of severe ARDS is complicated and strategies and protocols are desperately needed.

To achieve this goal, we will develop data-driven linear parameter-varying (LPV) dynamical systems models that relate patient clinical state and ventilator inputs to the output variable patient SpO2. Patient state will be characterized using data from the electronic health record (EHR) and minute-by-minute physiological time-series (PTS) data (e.g., heart rate, respiratory rate, SpO2) acquired from patient monitoring. We will first develop the LPV model using retrospective data from non-COVID-19 patients who are on ventilators to help treat conditions such as pneumonia and ARDS. Then, we will test the predictive capabilities of the LPV model in COVID-19 patients who are placed on ventilators. Finally, we will develop an optimal ventilator control strategy for COVID-19 patients to regulate SpO2 levels in ICU patients based on the LPV model. Attempting to control a complex biological system using control strategies based on mechanistic models is generally intractable. However, the LPV framework allows for sophisticated optimal strategies to be implemented that not only allow for better performance than other classical methods, but also provides stability and performance guarantees.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.